small ran seq illumina

[[email protected]]

Hello,

We just sequenced the banks you made from your NEBNext Small RNA kit.


On the other hand, it is totally impossible for us to analyze the data, you can see that the sequences contain almost only N after demultiplexing, while the quality of the sequencing run is perfect.

[GenoMax]

You are going to need to provide more information than that to get useful help. If the reads only contain N's after demultiplexing then there is not much you can do but start diagnosing what went wrong. First with the run and then work backwards all the way to the sample prep, if needed.

[[email protected]]

for the quality of the run it was very good,
but the fastQ file contains almost 90% of N
and no miRNA

[GenoMax]

What sequencer was this run on? What kind of chemistry was used? If 90% of your reads at N's then that is not a good run.

[[email protected]]

but the result of the run shows that it is of good quality it is a "miseq illumina"
with a NEB bank preparation

[[email protected]]

I just ran a MiSeq V3 150 cycles run (SR 50 bp only) with 12 libraries preparation with the "NEBNext Small RNA kit" from NEB. I also launched a adaptor trimming after the run (TruSeq adaptor).
Library profils are OK :


We have a problem because the run quality and demultiplexing are perfect to me.

But when we look at the fastQ files (for all of the samples), samples have approximatly 1.5M raw reads but almost 90% of the sequences are "N" with a very bad Qscore.

For example, I send you a sample fastQ file "86CorG1".
Post trimming : sequences are +/- 35 bp long (what we are expected).
Post filtration on a Q20 (to eliminate the "N" sequences) : 179.595 reads remained.
After analysis on BaseSpase : no miRNA detected.

Do you have an explanation for that? How is it possible that the quality run has a Q30 but the individual sample is at Q2-Q3 only?
Can the problem came from the adaptor trimming?

Thank you for your help,
Best,

[jdk787]

Take a look at this thread, it addresses the same issue you are having.

http://seqanswers.com/forums/showthread.php?t=70088

After trimming the adapters on the MiSeq you are left with short sequences that MiSeq Reporter masks with NNNs.

[GenoMax]

@josh: Thanks for the reminder.

@[email protected]: Re-demux the run without the adapter sequence in SampleSheet. You can then remove the adapter sequences with a different program. I recommend bbduk.sh from BBMap suite with literal=your_RNA_adapter_sequence option.